AI-Analysis-Assistant: Difference between revisions
mNo edit summary |
mNo edit summary |
||
| (2 intermediate revisions by the same user not shown) | |||
| Line 5: | Line 5: | ||
Reference file: [[Media: openFDA device_search_fields.xlsx]] | Reference file: [[Media: openFDA device_search_fields.xlsx]] | ||
{| role="presentation" class="wikitable mw-collapsible mw-collapsed" | {| role="presentation" style="min-width:250px;" class="wikitable mw-collapsible mw-collapsed code-sample" | ||
| <strong>Settings.txt</strong> | | <strong>Settings.txt</strong> | ||
|- | |- | ||
| Line 33: | Line 33: | ||
AI-ProblemSummaryPrompt: // Prompt that summarizes the product problem | AI-ProblemSummaryPrompt: // Prompt that summarizes the product problem | ||
AI-AnalysisPrompt: // Prompt that analyzes the MDR data to identify trends and patterns in adverse event occurrence | AI-AnalysisPrompt: // Prompt that analyzes the MDR data to identify trends and patterns in adverse event occurrence | ||
| Line 40: | Line 38: | ||
AI-ReportSummaryPrompt: // Prompt that summarizes the information collected from the MDR records | AI-ReportSummaryPrompt: // Prompt that summarizes the information collected from the MDR records | ||
AI-WordsPerReport: // Number of words in each intermediate report (recommended: | AI-WordsPerReport: // Number of words in each intermediate report (recommended: 10000, range: 100-30000) | ||
AI-ModelTemperature: // LLM Temperature index (recommended: 0.05, range: 0..1f) | AI-ModelTemperature: // LLM Temperature index (recommended: 0.05, range: 0..1f) | ||
AI-ModelTopP: // LLM TOP_P index (recommended: 0 | AI-ModelTopP: // LLM TOP_P index (recommended: 1.0, range: 0..1f) | ||
AI-ModelTopK: // LLM TOP_K number of words for next word prediction (recommended: | AI-ModelTopK: // LLM TOP_K number of words for next word prediction (recommended: 2, range: 1-128) | ||
AI-ModelMaxOutputTokens: // LLM maximum output words (range: 1- | AI-ModelMaxOutputTokens: // LLM maximum output words (range: 1-8192) | ||
|} | |} | ||
</blockquote> | </blockquote> | ||
| Line 58: | Line 56: | ||
==App Description== | ==App Description== | ||
'''Use | '''App Use Case:''' | ||
The AI-Analysis-Assistant app is a powerful tool designed to enhance decision-making and productivity by leveraging artificial intelligence and data mining of vast amounts of data, specifically focused on adverse events (MAUDE: Manufacturer and User Facility Device Experience). It supports a wide range of data analysis tasks using natural language and includes highly extensible options for analyzing unstructured data. From summarizing data analytics and trends to uncovering hidden patterns and insights, to simulating what-if scenarios with objective data-driven results, the Analysis-Assistant app enables users to consider multiple factors, analyze vast amounts of data efficiently, and make well-informed, bias-free decisions. | The AI-Analysis-Assistant app is a powerful tool designed to enhance decision-making and productivity by leveraging artificial intelligence and data mining of vast amounts of data, specifically focused on adverse events (MAUDE: Manufacturer and User Facility Device Experience). It supports a wide range of data analysis tasks using natural language and includes highly extensible options for analyzing unstructured data. From summarizing data analytics and trends to uncovering hidden patterns and insights, to simulating what-if scenarios with objective data-driven results, the Analysis-Assistant app enables users to consider multiple factors, analyze vast amounts of data efficiently, and make well-informed, bias-free decisions. | ||
''' | '''App Functionality:''' | ||
The AI-Analysis-Assistant app reads a Settings file that contains criteria for adverse events and pulls the up to date matching records from MAUDE. The unstructured data is then chunked to optimize semantic vector processing adopting a retrieval augmented generation approach. The model then performs the data analysis prompts given in the Settings file on the data chunked data and injects the results in a predefined report template. | The AI-Analysis-Assistant app reads a Settings file that contains criteria for adverse events and pulls the up to date matching records from MAUDE. The unstructured data is then chunked to optimize semantic vector processing adopting a retrieval augmented generation approach. The model then performs the data analysis prompts given in the Settings file on the data chunked data and injects the results in a predefined report template. | ||
Latest revision as of 23:04, 29 May 2024
App Files
Application executable: Media: AI-Analysis-Assistant.zip
Reference file: Media: openFDA device_search_fields.xlsx
Settings.txt CONFIG:OpenFDA-Vertex-V1.0 // identifies schema version (readonly)
https_url1: // openFDA api for source data
limit1: // limit of query results
report_title: // title of output report
data1_fields: // data fields to be analyzed in MDR query results
KEYWORD1-SEARCH: // general keyword search terms (matched in any field)
SEARCH1-FIELDS:SEARCH1-TERMS // query search fields and terms, see openFDA device_search_fields.xlsx for full details of searchable fields
date_received: // MAUDE example field
device.device_report_product_code: // MAUDE example field
SORT1-FIELD:SORT1-TERM // query results sorting date_received:desc
COUNT1-FIELD:COUNT1-TERM // query count field/term
AI-ProblemSummaryPrompt: // Prompt that summarizes the product problem
AI-AnalysisPrompt: // Prompt that analyzes the MDR data to identify trends and patterns in adverse event occurrence
AI-ReportSummaryPrompt: // Prompt that summarizes the information collected from the MDR records
AI-WordsPerReport: // Number of words in each intermediate report (recommended: 10000, range: 100-30000)
AI-ModelTemperature: // LLM Temperature index (recommended: 0.05, range: 0..1f)
AI-ModelTopP: // LLM TOP_P index (recommended: 1.0, range: 0..1f)
AI-ModelTopK: // LLM TOP_K number of words for next word prediction (recommended: 2, range: 1-128)
AI-ModelMaxOutputTokens: // LLM maximum output words (range: 1-8192)
See the sample files page for example settings and reports.
Sample Files: AI-Analysis-Assistant
App Description
App Use Case: The AI-Analysis-Assistant app is a powerful tool designed to enhance decision-making and productivity by leveraging artificial intelligence and data mining of vast amounts of data, specifically focused on adverse events (MAUDE: Manufacturer and User Facility Device Experience). It supports a wide range of data analysis tasks using natural language and includes highly extensible options for analyzing unstructured data. From summarizing data analytics and trends to uncovering hidden patterns and insights, to simulating what-if scenarios with objective data-driven results, the Analysis-Assistant app enables users to consider multiple factors, analyze vast amounts of data efficiently, and make well-informed, bias-free decisions.
App Functionality: The AI-Analysis-Assistant app reads a Settings file that contains criteria for adverse events and pulls the up to date matching records from MAUDE. The unstructured data is then chunked to optimize semantic vector processing adopting a retrieval augmented generation approach. The model then performs the data analysis prompts given in the Settings file on the data chunked data and injects the results in a predefined report template.
Diagram outlining the processing steps of the Analysis Assistant:
A PDF document is generated containing the analysis results.